In this proposal we seek to understand how human primary retinal pigment epithelial cells (RPE) differentiate into cells of the mesenchymal lineage. We have shown that RPE derived from a variety of sources, including human adult, fetal, ESC-derived, ARPE-19 and bovine, acquire mesenchymal phenotypes. Moreover, primary adult human RPE that were clonally expanded from a single cell produced adipocytes, chondrocytes and bone when placed in the respective differentiation media. This topic is important because RPE, while normally a stable epithelium, can undergo proliferative and metaplastic changes detrimental to vision. RPE metaplasia into mesenchymal-related cells has been long -associated with retinal disease, yet the mechanism underlying these changes remains a mystery. In aim 1 we will study how this plasticity is encoded at the level of the epigenome. Recent studies have identified epigenetic signatures at promoters and enhancers that indicate active genes and poised genes (those that could be activated if given appropriate cues). We will ask whether genes associated with mesenchymal lineages are in a poised conformation in human RPE cells, and whether they transit to an active form upon differentiation into osteo-, chondro- and adipocyte pathways. In aim 2 we will determine exogenous factors stimulating this differentiation, by testing factors known to act on mesenchymal stem cells (MSCs) during development towards osteogenic, chondrocyte and adipocyte fates, including members of the TGF-beta superfamily, and by use of small molecule inhibitors, which will help identify therapeutics to attenuate this process. One of the prominent abnormal cell fates associated with RPE metaplasia is ossification, hence it is particularly important to understand how RPE cells undergo osteogenesis. In aim 3 we will focus on the steps RPE move through when transitioning towards this fate, evaluating whether the transcription factor sequence known for normal osteogenesis generation is followed, or whether abnormal pathways are activated. Preliminary studies show dramatic upregulation of the essential osteogenic transcription factor Runx2 in RPE undergoing transformation towards bone. We will determine whether knocking down Runx2 using a shRNA approach will inhibit RPE osteogenesis. Together, these three aims address the central mechanisms, from exogenous factor through to gene regulation and epigenetic changes that underlie RPE phenotypic plasticity. Disruption of RPE stability is known to occur in several important retinal diseases such as epiretinal membrane formation, macular degeneration and phthsis bulbii, and understanding this plasticity will provide novel approaches toward developing treatments of these conditions. The findings will also have direct relevance to understanding the factors that maintain the RPE as a stable, pigmented, cobblestone polarized epithelial layer, which is especially timely given that RPE cells derived from human embryonic stem cells are entering clinical trials for retinal disease, and steps must be taken to minimize metaplastic changes to ensure development of safe transplantation conditions. PUBLIC HEALTH RELEVANCE: RPE metaplasia and acquisition of abnormal fates such as bone formation is associated with a number of retinal diseases such as proliferative vitreo-retinopathy and painful conditions such as phthisical eye. Here we will address the mechanism by which this occurs, from identifying the factors that cause these changes, to understanding how those factors alter gene expression. This study has direct relevance to RPE disease, but also has implications beyond the retina, to diseases affecting other epithelia that similarly can exhibit metaplastic changes associated with pathology.